Over the next five years, the Johns Hopkins pre-doctoral AHRQ NRSA training program will prepare PhD health services researchers within the following three high priority areas: 1) Reducing Health Care Disparities for Vulnerable Populations: This area will focus on racial and ethnic minorities (with a particular focus on African Americans), and populations disadvantaged by lack of access to health services (low income persons, uninsured children, residents of underserved areas) and vulnerable groups with complex care needs. 2) Improving Organization and Payment of Health Care Services: This area will focus on research that seeks to expand our understanding of the impact of organizational structure, provider payment, insurance coverage and design, and clinical and fiscal management on the efficiency, equity and quality of care. 3) Ensuring Patient Safety and Quality Health Care: This area will focus on the multi-dimensional factors associated with the attainment of patient safety and quality in diverse delivery settings. We will address clinical, management and patient perspectives. These three areas of pre-doctoral training build from a common core of course work that emphasize statistical, measurement, database/IT application, evaluation, and policy analysis skill sets. This proposal presents a detailed set of core competencies and learning objectives that will provide an important focus for the fellows' training. The competencies are based in large part on the recommendations of an AHRQ sponsored national "HSR core competencies" conference our Department hosted in 2005. Our training program benefits from a large faculty with great breadth and depth in the three priority areas and extensive research opportunities for students both at Johns Hopkins and in the Baltimore / Washington area. Our program will offer personalized small group and one-on-one support, mentoring and guidance to ensure a coordinated and cohesive training program experience for each of our fellows. [unreadable] [unreadable] [unreadable] [unreadable]